Pediatric Annals

Bacterial Meningitis in Childhood: Neurologic Complications and Their Management

Edgar Y Oppenheimer, MD; N Paul Rosman, MD

Abstract

INTRODUCTION

Before the introduction of chemotherapeutic and antibiotic agents, the outcome for patients with pyogenic meningitis was almost uniformly poor. Mortality varied from 60 to 100 per cent,1 and the survivors could be anticipated to have serious neurologic handicaps. The advent of antibiotics has dramatically reduced the mortality from meningitis, particularly that occurring after the neonatal period. While this recent progress has been very gratifying, the challenge facing today's clinician is perhaps greater than before, for the availability of modern antibiotic therapy has transformed a disease for which little could be done into one in which early intervention is imperative. Such treatment is necessary in order to maximize the patient's benefit from therapy and minimize the risk of complications, for, somewhat paradoxically, improved survival in bacterial meningitis has been accompanied by an increased number of complications from the disease. While further reduction of mortality continues to be an important goal of therapy, it is clear that the major clinical effort must be directed towards improvement in the quality of life for the survivors and, most important, prevention of the disease.

The purpose of this review is to discuss the neurologic complications of purulent meningitis in childhood, highlighting their preventable and treatable aspects.

SCOPE OF THE PROBLEM

The National Institutes of Health2 estimate that in the United States each year there are some 12,000 to 15,000 cases of Hentophilus influenzile meningitis, 4,500 to 5,000 cases of pneumococcal meningitis, and 4,500 to 5,000 cases of meningococcal meningitis in infants and young children. These three organisms account for about 70 per cent of all cases of bacterial meningitis beyond the neonatal period. Extrapolating from these estimates and taking an estimated mortality figure of 10 per cent (figures vary from 5 to 15 per cent), one might expect about 3,000 to 3,500 deaths from meningitis to occur each year in the United States. The frequency of H. influenzile meningitis - already the major cause of purulent meningitis in childhood, with 95 per cent of cases occurring in the first five years of life - may in fact be increasing further.3,4 It is estimated that at least one out of 2,000 children in the United States aged five years or less will contract the disease.5 Pneumococcal meningitis, the second most frequent cause of bacterial meningitis in children, is predominantly a disease of the very young and very old and is particularly likely to affect patients with impaired immunity or with chronic debilitating diseases. For example, one out of 27 patients with sickle cell disease may be expected to develop pneumococcal meningitis by age four years.5

The frequency of neurologic sequelae in survivors is more difficult to estimate, since the morbidity rates vary from 7 to 70 per cent, according to what degree of disability is considered significant and the length of follow-up. A study of children followed up after H. influenzae meningitis in Tennessee showed that 35 per cent of the survivors had severe or significant handicaps.6 Recent statistics reported by the National Institutes of Health2 show that between 30 and 50 per cent of infants and children who recover from H. influenzae type B meningitis suffer permanent disability, making this disease the leading cause of acquired mental retardation in the United States.

The occurrence rate for neonatal meningitis is 0.13 per 1,000 full-term births and is 17 times that number in prematures.7 The outcome for children who contract meningitis in the neonatal period continues to be poor, with a 30 to 65 per cent mortality and a considerable morbidity in the survivors. The outcome appears to be generally poorer with gram-negative infections, of which…

INTRODUCTION

Before the introduction of chemotherapeutic and antibiotic agents, the outcome for patients with pyogenic meningitis was almost uniformly poor. Mortality varied from 60 to 100 per cent,1 and the survivors could be anticipated to have serious neurologic handicaps. The advent of antibiotics has dramatically reduced the mortality from meningitis, particularly that occurring after the neonatal period. While this recent progress has been very gratifying, the challenge facing today's clinician is perhaps greater than before, for the availability of modern antibiotic therapy has transformed a disease for which little could be done into one in which early intervention is imperative. Such treatment is necessary in order to maximize the patient's benefit from therapy and minimize the risk of complications, for, somewhat paradoxically, improved survival in bacterial meningitis has been accompanied by an increased number of complications from the disease. While further reduction of mortality continues to be an important goal of therapy, it is clear that the major clinical effort must be directed towards improvement in the quality of life for the survivors and, most important, prevention of the disease.

The purpose of this review is to discuss the neurologic complications of purulent meningitis in childhood, highlighting their preventable and treatable aspects.

SCOPE OF THE PROBLEM

The National Institutes of Health2 estimate that in the United States each year there are some 12,000 to 15,000 cases of Hentophilus influenzile meningitis, 4,500 to 5,000 cases of pneumococcal meningitis, and 4,500 to 5,000 cases of meningococcal meningitis in infants and young children. These three organisms account for about 70 per cent of all cases of bacterial meningitis beyond the neonatal period. Extrapolating from these estimates and taking an estimated mortality figure of 10 per cent (figures vary from 5 to 15 per cent), one might expect about 3,000 to 3,500 deaths from meningitis to occur each year in the United States. The frequency of H. influenzile meningitis - already the major cause of purulent meningitis in childhood, with 95 per cent of cases occurring in the first five years of life - may in fact be increasing further.3,4 It is estimated that at least one out of 2,000 children in the United States aged five years or less will contract the disease.5 Pneumococcal meningitis, the second most frequent cause of bacterial meningitis in children, is predominantly a disease of the very young and very old and is particularly likely to affect patients with impaired immunity or with chronic debilitating diseases. For example, one out of 27 patients with sickle cell disease may be expected to develop pneumococcal meningitis by age four years.5

The frequency of neurologic sequelae in survivors is more difficult to estimate, since the morbidity rates vary from 7 to 70 per cent, according to what degree of disability is considered significant and the length of follow-up. A study of children followed up after H. influenzae meningitis in Tennessee showed that 35 per cent of the survivors had severe or significant handicaps.6 Recent statistics reported by the National Institutes of Health2 show that between 30 and 50 per cent of infants and children who recover from H. influenzae type B meningitis suffer permanent disability, making this disease the leading cause of acquired mental retardation in the United States.

The occurrence rate for neonatal meningitis is 0.13 per 1,000 full-term births and is 17 times that number in prematures.7 The outcome for children who contract meningitis in the neonatal period continues to be poor, with a 30 to 65 per cent mortality and a considerable morbidity in the survivors. The outcome appears to be generally poorer with gram-negative infections, of which Escherichia coli is the most common, than with grampositive organisms, such as Streptococcus, enterococcus, pneumococcus, and Listeria.8 Some of the differing results in published series of neonatal meningitis doubtlessly reflect different percentages of gram-negative and gram-positive organisms in the groups studied.

Clearly, while the therapeutic successes of the past several decades have greatly improved the overall outlook in purulent meningitis, significant mortality and morbidity for that disease continue to be problems of major proportions.

ANATOMIC AND PATHOLOGIC CONSIDERATIONS

The central nervous system (CNS) offers less resistance to infection than does any other body tissue; indeed, a number of organisms that are nonpathogenic or weakly pathogenic elsewhere become serious infective agents when they invade the CNS. It is thus fortunate that the CNS is shielded by bone and leptomeninges, and that the blood-brain barrier serves as an additional defense against infection9 (Figure 1). Once the invading organisms gain entry into the leptomeninges, they multiply rapidly within the cerebrospinal fluid (CSF). They then disseminate throughout the CSF pathways, inside the brain, within the ventricular system, and outside the brain, in the basal cisterns, over the cerebral and cerebellar convexities, and around the spinal cord. The leptomeninges provide an effective barrier against direct penetration of organisms into the brain substance; thus the parenchymal involvement seen in meningitis is not due to direct bacterial invasion of brain10 but, rather, is a consequence of inflammation of adjacent structures, particularly blood vessels.

"Toxic encephalopathy" and brain swelling characteristically develop during the acute phase of the illness. As the disease evolves, inflammation in the leptomeningeal vessels, particularly veins, can result in cerebral infarction.10 Concomitant inflammation within the subdural space can lead to an accumulation of sterile (effusion) or purulent (empyema) subdural fluid. In more longstanding infection - for example, when the diagnosis is delayed or therapy is inadequate - there is a fibroblastic proliferation in the leptomeninges. This can cause obstruction of CSF pathways, resulting in hydrocephalus and entrapment of cranial and peripheral nerves. This complication occurs with the greatest frequency following neonatal meningitis. On rare occasions, fibrous proliferation in the subarachnoid space can give rise to multiple subarachnoid cysts,11 which may cause progressive head enlargement and an initial misdiagnosis of hydrocephalus. Such leptomeningeal cysts can form reservoirs of persisting infection that are relatively inaccessible to systematically administered antibiotics.

CLINICAL CONSIDERATIONS

The clinical manifestations of meningitis and their sequelae develop over a considerable span of time, as would be anticipated from the anatomic and neuropathologic correlates outlined above. Some of these complications become clinically evident during the acute phase of the illness, while others evolve over a period of weeks or months following the acute illness (Table 1). The ensuing discussion will focus on some of the major neurologic sequelae of bacterial meningitis and their management.

RAISED INTRACRANIAL PRESSURE

Intracraneal hypertension is a very common finding in bacterial meningitis, as evidenced by the mean opening pressure on lumbar puncture, which is usually about 300 mm. of water.12 Contributing to this elevation in pressure are brain swelling, impaired circulation of CSF due to the purulent exúdate, and possibly increased CSF production due to inflammation in choroid plexuses.

Table

TABLE 1SUMMARY OF NEUROLOGIC COMPLICATIONS OF BACTERIAL MENINGITIS

TABLE 1

SUMMARY OF NEUROLOGIC COMPLICATIONS OF BACTERIAL MENINGITIS

Brain swelling is caused by several factors. In the acute phase of meningitis, the adjacent inflammatory reaction in the leptomeninges is thought to be responsible for the development of edema by increasing the permeability of vessels in the cerebral subarachnoid space and probably by causing the release of toxic factors that can penetrate into the brain substance and produce brain swelling.12 In addition, superadded hypoxia due to hypotension, prolonged seizures, hyperpyrexia, and depressed respirations can contribute to the cerebral edema. Brain swelling tends to be most severe in meningococcal meningitis.

In the uncomplicated case of meningitis, the acute brain swelling is a reversible process and will usually subside within 48 hours after institution of antibiotic therapy. Papilledema is usually not seen, probably because of the generally brief period of intracranial hypertension and, in infancy, because the open fontanelles and unfused cranial sutures prevent marked elevations of intracranial pressure. When papilledema is present, it should alert the physician to the possibility of coexisting pathology, such as subdural fluid collections, brain infarct, or brain abscess.

If the elevation of intracranial pressure is sufficiently great or occurs unequally within the cranial vault, brain herniation can result. Cerebellar herniation through the foramen magnum can cause compression of the medulla oblongata and compromise of vital cardiovascular and respiratory centers.12 Temporal lobe herniation can cause a third cranial nerve palsy by direct compression that is usually ipsilateral to the herniation. It can also cause an ipsilateral, contralateral, or bilateral hemiparesis by pressure against the tentorium or from direct compression. In neonatal meningitis, herniation is usually not seen, probably because the raised intracranial pressure in the newborn can be effectively accommodated by separation of the cranial sutures.13

Cerebral edema is often difficult to diagnose clinically, especially in a patient who is already ill and obtunded. Because of the importance of possible therapeutic intervention, it should be anticipated in any patient with meningitis with hypoxia, prolonged seizures, or persisting hyperpyrexia, and it should be considered to be present if the patient is unconscious or shows abnormal pupillary responses, elevated systolic blood pressure, decreased pulse rate, or irregular and/or slowed respirations. With both anticipated and evident cerebral edema, therapy to reduce intracranial hypertension should be instituted.14 For the more acute situation in which herniation is impending or is occurring, forced hyperventilation and mannitol (2-3 gm./kg.), urea (1-2 gm./kg.), orglycerol(0.5-1.5 gm./kg.), administered intravenously every four to six hours, may be used. In the less acute situation, glycerol and such corticosteroids as dexamethasone (0.4 mg./kg. stat. and 0.2 mg./kg. given intravenously every four to six hours) are the drugs most often employed.

In the clinical situation of a patient with raised intracranial pressure, there is invariably concern about the danger of precipitating herniation with a lumbar puncture. Whereas a lumbar puncture may accelerate death in a very few patients by producing herniation, that risk is far outweighed by the risk of failure to diagnose and appropriately treat a patient with meningitis. To lessen the risk of herniation in the face of elevated pressure, a small-bore lumbar puncture needle should be used, a small amount of fluid (no more than 2-3 ml.) should be removed slowly over several minutes, and a lumbar puncture should never be done with the patient in a sitting position. These measures are designed to prevent precipitous changes in pressure relationships within the CSF pathways and thereby to minimize the risk of herniation.

CRANIAL NERVE PALSIES

Cranial nerves 2, 3, 4, 6, 7, and 8 may be involved in bacterial meningitis, but ocular palsies, particularly those affecting the sixth cranial nerve, are the most frequent.

Ocular palsies. Sixth cranial nerve palsies are usually transient and can occur with raised intracranial pressure regardless of the cause. This nerve is particularly vulnerable, for it is the cranial nerve with the longest intracranial course. A sixth cranial nerve paralysis thus has no localizing value.

The third cranial nerve is the nerve most frequently affected in temporal lobe herniation. The parasympathetic pupillary fibers are located in the outermost portions of the nerve and are thus the ones usually compressed first by the extrinsic pressure; paiasympathetic paralysis results in unopposed sympathetic innervation to the pupil, and thus a dilated pupil is produced. When the pressure from the herniation is continued, the extraocular muscles innervated by the third nerve become paralyzed, resulting in an ophthahnoplegia with outward and downward deviation of the affected eye.

The importance of recognizing these ocular palsies is that they indicate a serious intracranial process that demands urgent treatment.

Auditory involvement. Deafness as a result of meningitis is a serious complication that is estimated to occur in about 3 per cent5 of all cases, but it has been reported to occur with frequencies as high as 20 per cent in meningococcal disease.15 Meningitic deafness is usually severe and almost always (95 per cent) bilateral. In a study of 1,362 cases of deafness due to infectious diseases reviewed in China, 337 (24 per cent) of the cases were related to meningitis.16 The deafness usually appears abruptly on the first or second day of the illness and rarely after the fourth day. In addition to the auditory involvement, about 80 per cent of patients also have a complete absence of vestibular function and about 20 per cent show diminished vestibular function on subsequent testing.16 Whereas the deafness is usually permanent and very troublesome, most patients can compensate adequately for their vestibular dysfunction within the first months following their meningitis. The mechanism by which auditory and vestibular functions are impaired is uncertain. It is possible that toxic substances are elaborated during the inflammatory processand cause damage to the eighth cranial nerve or the inner ear. Alternatively, there may be actual bacterial invasion of the inner ear via the cochlear aqueduct.17

Recent reports from Sweden have shown the frequency of severe sensorineural hearing loss in children with H. influenzae meningitis to be greater in children treated with ampicillin than in those treated with triple antibiotics (chloramphenicol, sulfonamides, and penicillin).18 These worrisome observations suggest that ampicillin in very high doses might be ototoxic or that, while ampicillin may be effective in the treatment of the meningitis, it may fail to protect the inner ear from the noxious effects of the infection.

Additional studies on the pathogenesis of deafness in bacterial meningitis are urgently needed so that approaches can be developed to minimize its occurrence.

ELECTROLYTE IMBALANCE AND WATER INTOXICATION

Electrolyte and water disturbances are common in patients with CNS infections. In the pediatrie age group the most common cause is probably iatrogenic, due to imprecise fluid management - especially in the small child, who is totally dependent on intravenous fluids during the acute phases of the illness.

In 1956 Nyhan and Cooke19 described the occurrence of hyponatremia, without dehydration or azotemia, in bacterial meningitis. This phenomenon has since been noted relatively often and is usually attributed to an inappropriate secretion of antidiuretic hormone (ADH). Such patients are unable to excrete a normal water load and thus become water intoxicated. The clinical manifestations of water intoxication, regardless of cause, include mild symptoms (headache, vomiting, and muscle cramps) or more serious ones (convulsions, cerebral edema, coma, and even death).15 Since many of these symptoms can occur in meningitis unaccompanied by inappropriate secretion of ADH, the diagnosis of the latter disorder is difficult on clinical grounds alone, but it can be established by measurement of the serum and urine sodium concentrations and osmolalities. The treatment of inappropriate secretion of ADH consists of fluid restriction to 1,000-1,200 ml./sq. m./day.

SEIZURES

Seizures have been reported to occur in 17 to 71 per cent of cases of bacterial meningitis, with an average frequency of about 25 to 30 per cent.20 In a retrospective review of the 207 cases of bacterial meningitis at the Massachusetts General Hospital studied by Swartz and Dodge,7,12 Rosman and co-workers20 determined that 60 per cent of seizures in this disease occurred within the first two days of the illness and none occurred after the eighth day. They also determined that when meningitis was complicated by seizures, the age-corrected mortality was almost doubled. This confirmed the findings of an earlier study by Ounsted,21 who had demonstrated that mortality and morbidity in meningitis were increased significantly when the illness was accompanied by seizures.

The seizures seen in meningitis are of three main types:5 brief, infrequent seizures that may be focal or generalized and usually do not constitute a management problem; recurrent, prolonged seizures with varying focality that tend to be difficult to control; and recurrent focal seizures. Although the appearance of focal seizures in bacterial meningitis would lead one to suspect definable focal pathology (such as a subdural effusion or an area of cerebral infarction) as the cause of these seizures, in 15 patients reviewed by Rosman et al.20 the seizures very frequently were not contralateral to the side of an associated brain infarct, abscess, subdural effusion, or otitis media.

These workers20 also showed that, regardless of the causative factors contributing to the production of seizures in purulent meningitis, the seizure frequency increased in direct proportion to the degree of temperature elevation. Indeed, fever appeared to be the only pathogenic factor that could be implicated in the cause of the seizures in 20 of 42 patients studied.20 It is possible that the poorer outlook in cases with associated seizures is artifactual, in the sense that patients with the highest temperatures may be the most seriill, with more severe cerebral damage, and thus are more likely to seizures. In such patients, the underlying brain pathology may well be the main determinant of the poorer prognosis. It is probable, however, that the seizures themselves, especially if prolonged, can be harmful to the brain by causing further hyperpyrexia, thereby increasing the brain's metabolic demands, or by causing superimposed hypoxic damage.

In view of these potential risks, every attempt should be made to prevent the occurrence of seizures in the meningitic patient. Careful monitoring of water and electrolyte status and prevention of high fever by adequate but not excessive hydration, sponging, and ace ty !salicylic acid) are extremely important in the prevention of seizures. In the severely ill patient, prophylactic phénobarbital (5-8 mg./kg./day given parenterally), especially during the early course of the disease, may be useful. If seizures do occur, it is imperative that careful attention be paid to the maintenance of an adequate airway, good oxygénation, normal blood pressure, and a balanced metabolic and acid-base status. The most helpful drugs for the acute management of seizures complicating bacterial meningitis are diazepam (Valium®) (0.3 mg./kg. given intravenously), phonobarbital (5-10 mg./kg. intravenously), and paraldehyde (1 ml. /3 kg. given intramuscularly or rectally).

SUBDURAL FLUID COLLECTIONS

Subdural effusions are said to occur in at least 50 per cent of children with meningitis,22 but their reported frequency appears to relate directly to the diligence with which they are sought. This would account for their reportedly greatest frequency in H. influenzile meningitis, for that bacterial agent causes most cases of meningitis in infancy, when the anterior fontanelle is still patent and a diagnostic subdural tap is thus more likely to be done. Of the large numbers of cases of meningitis with accompanying subdural effusions, probably only 10 to 20 per cent are symptomatic.5,12,19

The clinical diagnosis of subdural effusion is established if on a subdural tap one finds a collection of 2 ml. or more of fluid in the subdural space, a protein content in the fluid at least 40 mg./100 ml. greater than that in CSF obtained from the lumbar subarachnoid space at the same time, and a red blood cell count in the fluid of less than 1,000,000/cu. mm.23

If the subdural fluid is purulent, the collection is called a subdural empyema. This is a much rarer complication, and it almost always arises as a consequence of direct extension from a coexisting sinusitis or mastoiditis.

The diagnosis of a subdural effusion should be suspected in a child with meningitis who shows increased transulumination, with persistent or recurring fever after a bactériologie "cure/' if vomiting or other evidence of raised intracranial pressure develops, or if the child develops focal neurologic signs. Subdural effusions are rarely diagnosed after age two years, because at that age transillumination is usually not helpful and tapping the subdural spaces is more difficult, for it must be done through closely approximated skull bones or after burr holes have been placed.

The pathogenesis of subdural effusions is not completely understood. Investigators have emphasized the importance of inflammation in the subdural space, thought to coexist with that in the subarachnoid space. It is believed that this inflammation alters vascular permeability and thus allows transudation of fluid into the subdural space.12 Both newly formed "leaky" vessels in the wall of the outer membrane surrounding the subdural collection and the osmotic effect of the protein-rich fluid within the subdural space have been implicated in the causation of persistent subdural effusions in bacterial meningitis.

Treatment of subdural effusions is required only if there is evidence of increased intracranial pressure or related focal neurologic signs. Needle aspiration of the subdural space is usually satisfactory. When fluid is removed, no more than 10-15 ml. should be aspirated from each side at one time, for removal of excessive amounts of fluid may disturb intracranial pressure relationships and thereby promote intracranial (subdural) bleeding. Asymptomatic subdural effusions clear with the passage of time. There is no convincing evidence that surgical stripping of subdural membranes is needed or indeed desirable to prevent either reaccumulation of subdural fluid or subsequent restriction of brain growth. Benson and co-workers24 compared the frequency of resulting brain damage in pyogenic meningitis with and without subdural effusion and found no significant difference between the two groups. The presence of subdural fluid collections in bacterial meningitis may have little influence on the morbidity from the disease. If such patients do poorly, the outcome is probably related more directly to the severity of the meningitis and accompanying brain damage than to the associated accumulation of subdural fluid.

CORTICAL VEIN THROMBOSIS AND CEREBRAL INFARCTION

One of the earliest changes to occur in meningeal infection is infiltration of the walls of leptomeningeal veins and arteries with inflammatory cells. This can then cause vascular thromboses, particularly in veins. The predilection for thrombosis in veins, rather than arteries, is presumably due to the fact that vein walls are thinner and venous blood flow is slower. Venous thrombosis does not usually develop before the second week of the infection.

As a result of thrombosis, hemorrhagic infarction of brain can occur in the territory of the occluded vessel, giving rise to focal neurologic signs. For example, involvement of the superior sagittal sinus can result in a paraparesis or diparesis, while involvement of the middle cerebral vein may produce a contralateral hemiparesis. Seizures are a very common component of this clinical picture. The CSF in such cases is bloody or xanthochromic and under increased pressure.

Early treatment includes adequate hydration in an attempt to prevent further vascular- sludging, control of seizures, and lowering of the increased intracranial pressure. Later treatment, over months or years, may help to minimize subsequent motor disabilities. This frequently requires an intensive program of physical, occupational, and drug therapy, designed to meet the specific needs of the patient.

Figure 1. Sites of neurologic complications of bacterial meningitis. (Copyright 1953, 1972. CIBA Pharmaceutical Company. Division of CIBA-GEIGY Corporation. Reproduced, with permission , from The CIBA Collection of Medical Illustrations, by Frank H. Netter. M. D- All rights reserved.)

Figure 1. Sites of neurologic complications of bacterial meningitis. (Copyright 1953, 1972. CIBA Pharmaceutical Company. Division of CIBA-GEIGY Corporation. Reproduced, with permission , from The CIBA Collection of Medical Illustrations, by Frank H. Netter. M. D- All rights reserved.)

HYDROCEPHALUS

If the diagnosis of meningeal inflammation is delayed or the meningitis is ineffectively treated, proliferation of fibroblasts with deposition of collagen in the leptomeninges may occur, resulting in obstruction to the CSF pathways. When the meningeal thickening takes place around the basal cisterns, it can impede normal CSF flow both up over the cerebral convexities and down along the spinal subarachnoid space.

If the obstruction prevents outflow of CSF from the fourth ventricle, a noncommunicating hydrocephalus will result. A noncommunicating hydrocephalus can also result from severe inflammation within the ventricular system (ventriculitis). This usually occurs at the level of the aqueduct of Sylvius, which is the narrowest portion of the intracranial ventricular system (Figure 1). If adhesions occur around the ventral aspect of the pons and midbrain, preventing the flow of CSF upward to the middle cranial fossa, a communicating hydrocephalus is acquired. This is the more common type of hydrocephalus encountered following meningitis.12 Hydrocephalus occurs most frequently in the younger age groups. In an autopsy series of neonatal meningitis, Berman and Banker13 found that 14 of 25 newborns developed hydrocephalus (five of them noncommunicating and nine communicating).

Hydrocephalus tends to develop in the third or fourth week following meningitis and should be suspected if there is excessive head growth or other evidence of raised intracranial pressure,2 a "setting sun" sign in the eyes, development of paraparesis, or a generalized increase in transillumination of the cranial vault. Children should be re-examined periodically for at least three months following their recovery from meningitis, in a search for evidence of this complication. Early recognition of hydrocephalus is of particular importance, since the increased intracranial pressure can cause damage to the brain; this increase in pressure can lead to severe dissolution of brain tissue, particularly when brain damage has already occurred. The advent of computerized axial tomography25 in 1973 has proved to be an extremely useful means of establishing the diagnosis of hydrocephalus in clinically suspected cases, without the need to subject the child to an invasive neuroradiologic procedure.

Once the diagnosis of hydrocephalus is established, if the head continues to grow too quickly or there are other signs of unchecked intracranial hypertension, a neurosurgical shunting procedure is usually required. A ventriculoperitoneal shunt is generally preferred to a ventriculoatrial shunt, since the latter procedure usually requires more subsequent shunt revisions as the child grows. Also, in the presence of intracranial infection, a ventriculoatrial shunt may lead to generalized septicemia. Medical treatment with acetazolamide26 or isosorbide,27 drugs that slow CSF production, will sometimes suffice. In some cases of hydrocephalus following meningitis, intracranial compensatory mechanisms are apparently adequate to produce spontaneous arrest in the hydrocephalic process.

RADICULOPATHY

Since the cranial and spinal nerves are covered by pia and arachnoid, they would be expected to be affected by the inflammatory process seen in bacterial meningitis. Surprisingly, unless the infection is unusually severe or prolonged, direct nerve damage from the acute inflammation (excepting that to the eighth cranial nerve) is distinctly rare.10 Inflammation in the subarachnoid space can give rise to arachnoidal adhesions, however, and these can cause entrapment of motor and sensory nerves as they course through and out of the spinal canal, resulting in a variety of asymmetric neurologic deficits. Paresthesias and dysesthesias may be quite troublesome symptoms in these patients, and treatment aimed at relieving the discomfort may be necessary. Phenytoin sodium (Dilantin*) and carbamazepine (Tegretol®) have both been reasonably successful in this regard. Myelography serves to confirm a clinically suspected diagnosis of arachnoidal adhesions in such cases.

RECURRENT MENINGITIS

Repeated episodes of meningitis demand a careful search for an anatomic defect that can serve as a portal of entry for bacteria into the CSF.7-28 Such a defect may occur following a basal skull fracture, with damage to the cribriform plate. Occasionally, an anatomic defect such as a dermal sinus tract or a neurenteric cyst may be responsible. Parameningeal foci of infection - such as paranasal sinusitis, mastoiditis, or occasionally a brain abscess - may be the source of recurrent infection. The immunologie status of such patients should also be considered, for patients with immunoglobulin deficiencies or malignancies, patients receiving cytotoxic agents or corticosteroids, and postsplenectomy patients are particularly vulnerable to recurrent bacterial meningitis.

EPIDERMOID TUMOR

Repeated lumbar punctures may be complicated by the later development of an epidermoid tumor at the insertion site.29-30 This has been attributed to the introduction of epithelial cells into the spinal canal by the lumbar puncture needle/ and it is particularly likely to occur following lumbar puncture with a needle unprotected by an inner stylet. Treatment consists of decompression and, when possible, complete excision of the tumor.31

MENTAL RETARDATION

Because of difficulty in establishing the child's premorbid intellectual capacity, reliable data regarding the frequency of intellectual impairment following meningitis are considerably more difficult to obtain than data concerning such sequelae as motor disabilities.

In a study by Hinton32 of 1,136 children with mental retardation, the mental subnormality was related to postnatally acquired meningitis in 11 per cent. Sell and co-workers6 carried out a controlled study of children recovering from H. influenzae meningitis, using a battery of psychologic tests to evaluate the postmeningitic children along with age-matched nonmeningitic siblings and classroom peers. Thirty-five per cent of the survivors were found to have severe or significant neurologic handicaps, and 16 per cent had possible CNS defects. A psychologic study carried out by Wolff33 on 134 patients who had recovered from meningococcal meningitis is of interest. Meningitis in the first six months of life was found to have the most profound effect on intelligence, for most children in this group scored below the average range and 20 per cent were found to be mentally retarded. In children six to 12 months old at the time of their meningitis, an adverse effect on intelligence was less discernible and severe mental retardation was uncommon. In only very few of the children in whom the disease occurred after the age of one year did impairment of intelligence result.

These observations underscore the need, particularly in the young child following meningitis, for careful neurologic and psychologic evaluation. It seems apparent that just as there is a continuum of motor abnormalities following purulent meningitis, there is also a spectrum of intellectual disabilities among the surviving patients. It is probable that careful examination of patients over a prolonged follow-up period would uncover a large number of cases of previously unsuspected minor cerebral dysfunction and behavioral abnormality.

PREVENTION OF MENINGITIS

Clearly, the neurologic sequelae of meningitis will not be totally eradicated until the disease can be prevented. Towards that end, vaccines have been developed in the past several years against pneumococci, H. influenzae type B, and group A and C meningococá. These vaccines, extracted from highly purified capsular polysaccharides, are relatively simple and evoke few local or systemic reactions. The levels of antibody response produced by these vaccines appear to be age-related, with children over 18 months of age showing just one-half the antibody levels produced in the adult, while the infant achieves only 1/20 to 1/50 of the adult levels.34 The crucial question now being investigated is whether the low antibody levels attained in infants and children are sufficient to provide adequate protection from infection. The results of early studies seem encouraging, but a great deal more investigation will be necessary before these vaccines can be recommended for routine clinical application in children.

CONCLUSION

The mortality and morbidity from bacterial meningitis continue to be significant problems, despite development of many new antibiotics and other therapeutic refinements. It thus remains of paramount importance, for a favorable outcome to the patient, that the clinician diagnose meningeal infection early, institute treatment promptly, and develop an understanding and anticipation of the potential complications of the disease. Meticulous attention to the total medical management of the patient must continue well beyond the period of hospitaliza ti on, to ensure early recognition and treatment of the many later-developing neurologic disabilities. Further major advances in the management of this serious illness await development of effective means of prophylaxis.

BIBLIOGRAPHY

1. Hutchison, P. A., and Kovacs, M. C. The sequelae of acute purulent meningitis in childhood Can. Med. Assoc. J. 89 (1973), 158.

2. National Institutes of Health, U.S. Public Health Service, Department of Health, Education, and Welfare. Recent Advances. Publication No. (NIH) 75-3, 1975, p. 45.

3. Michaels. R. H. Increase in influenzai meningitis. N. Engl. J. Med. 285 (1971), 666.

4. Smith, E. W., Jr., and Haynes R. E. Changing incidence of Hemophilus influenzae meningitis. Pediatrics 50 (1972), 723.

5. Smith, D. H,, el al. Bacterial meningitis- A symposium. Pediatrics 52 (1973), 586.

6. Sell, S. H., et al. Long-term sequelae of Hemophilus influenzae meningitis. Pediatrics 49 (1972), 206.

7. Swartz. M. N., and Dodge, P. R. Bacterial meningitis - A review of selected aspects. I General clinical features, special problems and unusual meningeal reactions mimicking bacterial meningitis. N. Engl. J. Med- 272 (1965), 725, 779, 842, 89T.

8. Fitzhardinge, P. M., et al. Long-term sequelae of neonatal meningitis. Dev. Med. Child Neurol. 16,, (1974). 3.

9. Greenfield, J. G. Neuropathology, Second Edition. Baltimore. The Williams & Wilkins Company, 1963, ch. 3, p. 138.

10. Adams, R. D., Kubik, C. S., and Bonner, F. J. The clinical and pathological aspects of influenzai meningitis. Arch. Pediatr. 65 (1948), 354, 408.

11. Mukherjee, S. K., Heilman, K. M , and Rosrnan, N. P. An unusual cause of progressive head enlargement following Proteus mirabilis meningitis. Dev. Med. Child Neurol. 11 (1969), 633.

12. Dodge, P, R.. and Swartz, M. N Bacterial meningitis - A review of selected aspects. II: Special neurologic problems, postmeningitic complications and clinicopathological correlations, N. Engl. J. Med. 272(1965), 954. 1003.

13. Berman, P. H., and Banker. B. Q. Neonatal meningitis: A clinical and pathological study of 29 cases. Pediatrics 38 (1966), 6.

14. Rosman, N P. Increased intracranial pressure in childhood, Pediatr Clin. North Am. 21 (1974), 483.

15. Nyhan, W. L,, and Richardson, F. Complications of meningitis. Annu, Rev. Med. 14 (1963), 243.

16. Teng.Y.C., Uu, J. H, and Hsu. Y. H. Meningitis and deafness: Repon· of 337 cases ol deafness due to cerebrospinal meningitis. Chin. Med. J. 81 (1962), 127.

17. lgarashi M,, and Schuknecht, H, F. Pneumococcic otitis media, meningitis and labyrinthitis. Arch. Otolaryngol. 76 (1962), 126.

18. Gamstorp, I,, and Ktockhoff, I. Bilateral, severe, sensori-neural hearing loss after Haemophilus influenzae meningitis in childhood, Neuropaediatric 5 (1974), 121.

19. Nyhan, W. L., and Cooke, R. E. Symptomatic hyponatremia in acute infections of central nervous system. Pediatrics 18 (1956), 604.

20. Rosman, N. P., Peterson, D. B., and Gölten. T. Seizures in bacterial meningitis: Patterns, pathogenesis, prognosis. Proceedings. First National Meeting, Child Neurology Society, Ann Arbor. Michigan, 1972, p. 12.

21. Ounsted, C. Significance of convulsions In children with purulent meningitis. Lancet J (1951), 1245.

22. Goodman, J. M., and Meatey, J. Postmeningitic subdural effusions: The syndrome and its management. J. Neurosurg. 30 (1969), 658.

23. Rabe, E. F., Flynn, R. E.. and Dodge, P. R. Subdural collections of fluid in infants and children. Neurology 18 (1968), 559.

24. Benson, P., Nyhan, W. L., and Shimizu. H. Prognosis of subdural effusions complicating pyogenic meningitis. J. Pediatr. 57 (960), 670.

25. Harwood-Nash, D. C., FiIz1 C. R., and Reilly. B. J. Cranial computed tomography in infants and children. Can. Med. Assoc. J. 113 (1975), 546.

26. Huttenlocher, P. R. Treatment of hydrocephalus with acetazolamide. J. Pediatr. 66 (1965), 1023.

27. Shurtlefi. D, B., et al. Temporary treatment of hydrocephalus and myelodysplasia with isosorbide: Preliminary report. J. Pediatr. 84 (1973), 651.

28. Humphreys, R. P. Surgical management of bacterial meningitis. Can. Med. Assoc. J. 113 (1975), 536.

29. Choremis. C., et al. lntraspmal eptdermoid tumours (choie steal ornas) in patients treated for tuberculous meningitis. Lancet 2 (1956), 437.

30. Sahs, A. L. Complications and sequelae of pyogenic meningitis. Clin. Neurosurg. 14 (1966), 211.

31. Muider, D. W., and Dale, A. J. D. Spinal cord tumors and disks. In Baker, A. B., and Baker, L. H. (eds.). Clinical Neurology, Second Edition, Volume III. New York: Harper & Row, 1975. ch. 33.

32. Hinton, G. G. Postnatal organic causes of mental retardation. Can. Med. Assoc. J, 87 (1962), 501.

33. Wolff, O. H. Proceedings, 23rd General Meeting, British Paediatric Association. Arch. Dis. Child. 27 (1952), 302.

34. Gold, R. Bacterial meningitis: Current status of vaccines against the meningococcus and Hemophilus influenzae type B. Conn. Med. 38 (1974), 304.

TABLE 1

SUMMARY OF NEUROLOGIC COMPLICATIONS OF BACTERIAL MENINGITIS

10.3928/0090-4481-19760501-04

Sign up to receive

Journal E-contents